Geometric and colour data fusion for outdoor 3D models.

Area of Science:

• Computer Vision
• 3D Reconstruction
• Geospatial Technology

Background:

• Generating accurate 3D models of outdoor scenes from scanners faces challenges due to environmental conditions and technical limitations.
• Existing methods for 3D model creation can be inefficient, particularly when dealing with varying illumination and scanner-specific cameras.

Purpose of the Study:

• To develop an efficient and robust method for generating accurate, dense, and colored 3D models of outdoor environments.
• To integrate external image data with scanner-acquired range data for enhanced photorealistic 3D reconstruction.
• To decouple color integration and geometry reconstruction for improved control and flexibility.

Main Methods:

• Proposed a novel approach for 3D model generation by integrating images and range data.
• Decoupled color integration and geometry reconstruction into independent processes.
• Developed a method to modify existing 3D model textures with external image color information.
• Created an algorithm for direct assignment of external images onto geometric models, bypassing online calibration.

Main Results:

• Successfully generated accurate, dense, and colored 3D models of outdoor archaeological sites.
• Demonstrated the effectiveness of integrating external images for improved texture quality in 3D models.
• Showcased the avoidance of tedious online calibration processes through direct image assignment.

Conclusions:

• The proposed approach enhances the accuracy, density, and color quality of 3D models for outdoor scenarios.
• This method offers a more efficient and flexible solution for 3D digitalization and modeling, particularly in challenging environments.
• The technique is applicable to fields such as cultural heritage preservation and digital archaeology.